#ResearchSpeak: grand challenges in MS (7.1) – working out how anti-B cell therapies work

How do B cell depleting agents work? Will they prevent SPMS? #ResearchSpeak #MSBlog #MSResearch


“It is now clear that B cells play a central role in pathogenesis of MS:


1. MS has not be seen in people who don’t make immunoglobulins due to rare genetic disorders (agammaglobulinaemia).



This observation is based on a literature review and my own personal experience. If anyone reading this post has seen a case of MS occurring in a patient with agammaglobulinaemia please let me know. 

2. Oligoclonal IgG bands are invariably found in the cerebrospinal fluid (CSF) and brain/spinal tissues of MSers. These immunoglobulin bands have been highly selected; in other words whatever is driving their production is getting help from other cells (antigen presenting cells and T cells). This selection process is typically seen with infections or foreign antigens (proteins).

I have always hypothesised that if we can uncover what these bands are reacting with we will uncover, or pin down, the cause of MS. 


MSers who don’t have these oligoclonal bands have a more benign course; this applies to MSers with relapse-onset MS and PPMS.

OCB-negative MSers don’t have ‘classic MS’ and should be given a separate diagnosis or the diagnosis of definitive MS should be delayed; this is particularly important when diagnosing PPMS! This is another reason why lumbar punctures and spinal fluid analysis should be done in all MSers. Please note to get into the ocrelizumab PPMS trial (Oratorio Trial) you had to have OCBs. The latter may be the reason why the trial was positive.”


You only get one chance not to make the diagnosis of MS and that is in the beginning or the diagnostic phase of the disease. Spinal fluid analysis helps you exclude other conditions and tells you if you have OCBs or not.


3. B-cell follicles, were the B-cells mature and become antibody factories or plasma cells, are found in CNS of MSers. These are ectopic follicles as they are normally found in lymph nodes and the spleen. These follicles appear to be more common in progressive MSers and may result in cortical or gray matter pathology that drives progressive disease. These follicles also contain plasma cells that don’t express CD20 the antigen targeted by ocrelizumab. 


The lack of targeting of plasma cells may prove to be the achilles heel of the anti-CD20 therapies. Killing plasma cells may be very important to prevent the SPMS. We will need to wait and see what happens with the long term follow-up of the ocrelizumab trial cohorts. 

4. Pathological studies show immunoglobulin deposits in the brains of MSers and complement activation. Complement is one of the molecules that certain classes of immunoglobulins use to damage or kill target cells and organisms. Interestingly in the, now famous, Barnett and Prineus lesion immunoglubluin deposition was seen without the classic T cell infiltrates. Many of us now consider this to be the earliest relapse-causing lesion and based on the observations in these lesions immunoglobulins seem to be is very important inflammatory mediators.

5. B cells are the cells were EBV, the virus that is causally linked to MS, resides.

Could the B cell be the Trojan horse that takes EBV into the brain and spinal cord? Unfortunately, the evidence on this topic is rather mixed at present, but I suspect the B-cell is the cell that EBV hitches a ride in. This may be the mechanism of how ocrelizumab works.  


6. Most if not all highly effective MS DMTs target B cells.

This is an observation that I made years ago! The one exception to that rule is daclizumab (anti-CD25). This is why we need to study the immunological effects of daclizumab in great detail in MS. We need to know exactly how it is working and what its effects are on B cell function. I have hypothesised in the past that the expansion of CD56-bright NK cells, that occurs with daclizumab, may be working via an antiviral mechanism.  


7. Targeted anti-B cell therapies (anti-CD20 – rituximab, ocrelizumab, ofatumumab) are among the most promising emerging MS therapies. There superior efficacy was not expected given the current dogma that MS is T-cell driven disease.

It is a pity that anti-CD20 therapies do not target plasma cells or plasmablasts; these are the immunoglobulin factory cells and are also part of immunoglobulin pathway. May be the newer anti-CD19 agents that target plasma blasts will be more effective than anti-CD20?

The Grand Challenge: Are B cells the pivotal cell in the pathogenesis of MS? How do anti-B cell therapies work in MS? Do anti-B cell therapies work as anti-EBV agents? How do we reconcile the immunological effects of  daclizumab with the B cell hypothesis? 


CoI: multiple

24 thoughts on “#ResearchSpeak: grand challenges in MS (7.1) – working out how anti-B cell therapies work”

  1. Why is it so difficult to find out what the oligoclonal bands are interacting with? How have they figured out what other other immunoglobulins (in other diseases) bind to? Also, is the development of B-cell follicles in the brain the event that triggers progression?

  2. Agammaglobulinemia is rare ( 1:390 000) for X-linked agammaglobulinemia, so is MS ( not in an MS clinic, but generally it is something like 1:400-1000 people). Say we take the 1:1000 for ease of calculation – there would be a 1:390 000 000 chance of both occuring together?

  3. "MSers who don't have these oligoclonal bands have a more benign course; this applies to MSers with relapse-onset MS and PPMS."I'm interested in this because I don't have OCB's yet my EDSS is 4. I would be interested in reading the studies that made you come to this conclusion because I could not find any data that confirms your hypothesesis.

    1. On an population level there may be the correlation but on the individual level it means nothing as there are always exceptions. This is because the correlations are weak and not absolute.

  4. This is very interesting. Can I ask a couple of questions please? It makes sense that the B cells are like 'command and control' nodes (or perhaps just initial command) in an air defense network. Are B cells themselves 'armed'? Do they have any capability to attack on their own? Do we know anything about the evolution of the immune system from simple creatures; did B cells evolve first or T cells?

  5. Ocrelizumab and other anti-CD20's are antibodies, so they don't easily get into the CNS. TB cells they target are residing in the periphery, and perhaps infiltrate into the CNS (through the "leaky" blood-brain barrier?). At the same time, OCB's result from endogenous plasma cell-secreted antibodies within the CNS. So, different B-cells are responsible for OCB's. How to reconcile these observations?Also, how would anti-CD19 get into the brain?

    1. Antibodies come from plasma cells and these are not inhibited by either anti-CD20 or anti-CD19. Anti-CD19 data was presented at ECTRIMS

    2. I keep writing this, but doesn't appear to be posting (and can't think of any reason why you'd want block it), so am assuming a technical snag. If not, let me know! :)What about Elotuzumab? Anti cd-319? Currently in Phase III for myeloma, pretty well targeted at mature plasma cells.Why don't you guys run a small proof of concept trial and see if it wipes out OCBs?Since the pharma company who produce it aren't in the MS game, presumably they'd be up for giving you the drug for free on a small-scale proof of concept trial… It's already passed Phase I and II for safety, and there's a highly measurable and relatively quick outcome in terms of impact on OCBs, which would no doubt make Daily Mail headlines as "cure of the week" if successful – which is good for business.Surely, after all these years of studies pointing the finger at some involvement of plasma cells it's now time to put it to the test? Especially when there are so many pre-approved and highly targeted anti-plasma cell treatments in the oncology world, both in development and ready to go…If you could have a safe infusion of a mab that got rid of OCBs, what MS'er wouldn't take it?

    3. I have not found anything in spamElotuzumab, an monoclonal antibody targeting signaling lymphocytic activation molecule F7 (SLAMF7) which is indeed found on plasma cells. It looks to be an IgG which acts via Natural killer cell activation to kill the target rather than complement fixation. However there are other variants.The antibody is made I believe by Abbvie, who are is actually into MS with their daclizumab and so maybe an idea ProfG. I'll email him

    1. Thanks MD,If we join the protein post with this thread, then does the EBV/B cell behave like a Trojan horse to get some protein into the CNS and then the CNS do the rest?

  6. 'The now famous Barnett and Prineas lesion…immunoglobulin deposition.seen without T cell infiltrates'. Seven years before this paper, the first description of initial primary MS lesions was reported in tissues from exceptionally early cases.(Brain,1997,120:1461-83) The cardinal feature of these lesions was the detection of Immune complexes of IgG and C3d complement on activated microglia, without T cell infiltration or BBB leakage. Subsequent studies on these early tissues extended this data and confirmed the observation that these immune complexes ( containing some extrinsic antigen) occur in tissues without evidence of anti myelin or any other parenchymal immune target. (Clin Neurol Neurosurg 108(2006) 234-240. Considering alone the consequences of microglial activation by these complexes, and the nature of the antigen(s) inducing them, You need to read and reflect on these papers which have obviously passed you by!

Leave a Reply to MouseDoctorCancel reply

Discover more from Prof G's MS Blog Archive

Subscribe now to keep reading and get access to the full archive.

Continue reading